Download AN IMPROVED ANALYSIS OF DATA MINING TECHNIQUES ON

M. Chandamona et al.,
Available Online http://www.ijncse.com
ISSN Online: 2395-7018
Special issue ,March, 1-9
AN IMPROVED ANALYSIS OF DATA MINING TECHNIQUES ON MEDICAL
DATA
Mrs.M.Chandamona M.Sc.,M.Phil,
Assistant Professor & Head of the Department BCA
G.T.N Arts College, Karur Road
Dindigul
Dr.PonPeriasamy,M.Sc.,MCA.,M.Phil.,
Associate Professor
Nehru Memorial College, Puthanampatti,
Trichy
1.INTRODUCTION
Abstract
In the last decade there has been increasing
usage of data mining techniques on medical data for
locating helpful trends or patterns that are utilized in
identification and higher cognitive process. data mining
techniques like cluster, classification, regression,
association rule mining, CART (Classification and
Regression Tree) area unit wide utilized in attention
domain. data mining algorithms, once appropriately
used, are capable of rising the standard of prediction,
identification and un wellness classification. the most
focus of this paper is to research data mining techniques
needed for medical data mining particularly to find
locally frequent diseases like heart ailments, lung
cancer, breast cancer and so on. we tend to measure the
data mining techniques for finding locally frequent
patterns in terms of value, performance, speed and
accuracy. we have a tendency to additionally compare
data mining techniques with typical strategies.
Keywords—Data mining, frequent patterns, data mining
techniques, medical data mining
Data mining is the method of excavation data for
locating latent patterns which may be translated into
valuable info. data mining usage witnessed
unprecedented growth within the previous couple of
years. currently the quality of data mining techniques
has been accomplished in health care domain. This
realization is within the wake of explosion of
complicated medical data. Medical data mining will
exploit the hidden patterns present in voluminous
medical data which otherwise is left undiscovered.
data mining techniques that are applied to medical
data embrace association rule mining for locating
frequent patterns, prediction, classification and
agglomeration. historically data processing techniques
were utilized in numerous domains. However, it's
introduced comparatively late into the health care
domain. however, as on nowadays lot of analysis is
found within the literature. This has LED to the
development of intelligent systems and decision
support systems in health care domain for correct
diagnosing of diseases, predicting the severity of
varied diseases, and remote health observation.
particularly the data mining techniques are a lot of
helpful in predicting heart diseases, lung cancer, and
breast cancer and so on.
The data mining techniques that are applied to
medical data embrace Apriori and FPGrowth [1], [2],
Int J Nano Corr Sci and Engg 3(3) (2016) 85-90
Editors: Dr S Rajendran, Dr M Pandiarajan, and A. Daniel Das
85
M. Chandamona et al.,
[3], [4], [5], [6], [7], and [8], unattended neural
networks [9][10], linear genetic programming [9],
Association rule mining [11], [12], Bayesian Ying
yang [13], decision tree algorithms like ID3, C4.5, C5,
and CART [14], [15], [16], [17], [18], [19], [20],
outlier prediction technique [21], Fuzzy cluster
analysis [22], classification algorithmic rule [17],
[23], [24], Bayesian Network algorithmic rule [14],
[25], Naive Bayesian [26], combination of K-means,
Self-Organizing Map (SOM) and Naïve Bayes [27],
statistic technique [28], [29], combination of SVM,
ANN and ID3 [16], agglomeration and classification
[30],SVM [16], [31], FCM [29],k-NN [24], and
Bayesian Network [14]. This paper provides the
outline of these techniques in terms of the matter they
solve or their utility in medical data processing or the
tools that are enforced over them and so on.
The remainder of the paper is structured as follows.
Section II reviews literature bearing on data mining
and applications of data mining techniques in health
care domain. Section III provides the outline of
medical data mining techniques. Section IV presents
the importance of discovering locally frequent
patterns or trends or diseases in medical knowledge.
Section V concludes the paper.
RELATED WORK
Data mining could be a process of analyzing
voluminous data in numerous views so as to motivate
trends or patterns that cause business intelligence [32].
data mining plays a vital role in IT because it
discovers data from historical data of varied domains.
for example data mining may be wont to mine
medical data as health care domain produces large
amount of information concerning patients, diseases,
diagnosis, medication and so on. By applying data
mining techniques in health care domain, the directors
will improve the QoS (Quality of Service) by
discovering latent probably helpful trends needed by
diagnosing [33]. data mining is helpful in medical
applications like medications, medical tests,
prediction of surgical procedures, and discovery of
relationships between pathological data and clinical
data [34]. Apriori and FPGrowth are the most widely
used frequent pattern mining algorithms [35]. These 2
algorithms and algorithms supported them are studied
in [2], [3], [4], [5], [6], [7], and [8]. These 2
algorithms are utilized in medical data mining.
Goodwin et al. [36] applied data mining techniques
for birth outcomes. Evans et al. [37] expressed that
hereditary syndromes will be detected automatically
using data mining techniques. Doron Shalvi and St.
Nicholas DeClaris, [10] mentioned medical data
mining through unsupervised neural networks besides
a technique for data visual image. They additionally
stressed the necessity for preprocessing before
medical data mining. within the year 2000 Krzysztof
J. Cior [38], biotechnology academician, known the
requirement for data mining methods to mine medical
multimedia system content. Tsumoto [39] known
issues in medical data mining. the problems include
missing values, data storage with reference to
temporal information and ambiguous data, completely
different medical coding systems being used in
Hospital Info Systems (HIS). Brameier and Banzhaf
[9] explored and analyzed 2 programming models like
neural networks, and linier genetic programming for
medical data mining. Abidi and Hoe [40] projected
and enforced a symbolic rule extraction work table for
generating emerging rule-sets. Abidi et al. [41]
explored the usage of rule-sets as results of
information mining for building rule-based
professional systems. Olukunle and Ehikioya [11]
proposed an algorithmic rule for extracting
association rules from medical image data. The
association rule mining discovers frequently occurring
things within the given dataset. wedge and Xu [13]
proposed a classification methodology supported
Bayesian Ying rule (BYY) that may be a 3 stratified
model.
They applied this model to classify liver disease
through automatic discovery of medical trends. Brunie
et al. [42] proposed architecture for mining genomedical data in heterogeneous and grid-based
distributed infrastructures. Mahmud Khan et al. [15]
targeted on call tree data mining algorithmic rule for
medical image analysis. particularly they studied on
lung cancer diagnosing through classification of x-ray
images. Podgorelec et al. [21] given an outlier
prediction methodology for rising performance of
classification as a part of medical data mining. Wang
et al. [22] applied fuzzy cluster analysis for medical
images. They used decision tree rule to classify
diagnostic procedure into traditional and abnormal
cases. Cheng et al. [17] applied classification
algorithmic rule to diagnose cardio vascular diseases.
For classification effectiveness they targeted on 2
feature extraction techniques particularly automatic
feature choice and professional judgment. Seng et al.
[43] introduced web based data mining for the
applying of telemedicine. Ghannad-Rezaie et al. [44]
conferred an approach to integrate PSO rule mining
strategies and classifier on patient dataset. They used
Particle Swarm optimisation technique moreover.
The results discovered that, their approach is
capable of performing arts surgery candidate selection
method effectively in brain disease. bethel et al. [12]
developed an association rule learner that is
predicated on the standards collected from past breast
cancer patients. The rule learner is employed during a
tool by name “Clinical Trial Assignment Professional
System”. Xue et al. [25] proposed and applied
Int J Nano Corr Sci and Engg 3(3) (2016) 85-90
Editors: Dr S Rajendran, Dr M Pandiarajan, and A. Daniel Das
86
M. Chandamona et al.,
Bayesian Network algorithmic rule for diagnosing of
an disorder called Coronary cardiopathy (CHD).
Abraham et al. [26] projected discrimination
techniques to enhance the accuracy of classification of
medical data victimisation Naive Bayesian classifier
algorithmic rule. Hassan and Verma [27] projected a
hybrid approach for classification of medical data
which mixes K-means, Self-Organizing Map (SOM)
and Naïve Thomas Bayes with NN based mostly
classifier. Tsumoto [45] studied multi-stage
diagnosing victimization experts’ diagnostic rules and
diagnostic taxonomy. They targeted on automatic
grouping of medical data extracted from clinical
database. Berlingerio et al. [28] studied Time
Annotated Sequences (TAS) rule for mining medical
knowledge with temporal dimensions. The extracted
patterns exhibited the attribute relationships in time
domain that helps in correct identification. Xing et al.
[16] developed data mining techniques for predicting
the chance of survival of CHD patients. to realize this
they combined 3 prediction models like SVM
(Support Vector Machine), Artificial Neural Networks
(ANN), and call trees victimisation C4.5 or ID3,
CART and C5. Abe et al. [46] projected an integrated
time-series data mining environment for mining large
amount of medical data for extracting additional
valuable rule-sets. Jiquan et al. [47] projected a
framework referred to as term-mapping to mix
multiple medical data sources for data processing.
Barnathan et al. [30] conferred a framework for
cluster, classification and similarity search of
biomedical images or 2d and 3D in nature. Shusaku et
al. [48] projected multi-scale matching and clump
technique on medical knowledge.
Their results unconcealed that their technique is
capable of grouping liver disease data supported
temporal covariance of choline esterase, albumin and
living substance. Hai Wang, and Shouhong Wang
[49] studied on the role of doctors in medical data
mining. medical experts will provide expert
recommendation that may be used as input in medical
data mining. Abdullah et al. [1] applied apriori rule
for medical data mining. They extracted frequent item
sets by analyzing associations between treatments and
diagnosis. Saraee et al. [18] applied data mining
techniques to medical data referring to military with
reference to death rate in kids because of accidents.
They used CART algorithmic rule to get a decision
tree. Balakrishnan and Narayanaswamy [31]
conferred feature choice victimisation SVM for
classifying diabetes databases. medicine and health
effects are well-mined by Froelich and Wakulicz-Deja
[29] victimization adaptive FCM (Fuzzy cognitive
Maps). Their work has LED to improved call support
and designing in health care domain. known as
Knowledge Discover Question Language for
preparing questions that are used to discover
knowledge from medical data. They explored ways
and means for intelligent medical data mining.
SUMMARY OF TECHNIQUES FOR MEDICAL
DATA MINING
Data mining techniques have shown significant
improvement in medical industry in terms of
prediction and decision making with respect to
various diseases like cancer, cardio vascular
abnormalities, diabetes, and others. Table 1 summary
the medical data mining, its areas of application and
the utility of the techniques.
CONCLUSION
In this paper we tend to survey numerous
data mining techniques that are utilized for medical
data mining. data mining techniques have higher
utility in medical data mining as there's voluminous
data during this industry. Attributable to the rising of
medical data, it's become indispensable to use data
mining techniques to assist decision support and
predication systems within the field of health care.
The medical mining yields required business
intelligence to support well informed diagnosing and
decisions. This paper has provided the outline of data
mining techniques used for medical data mining
besides the diseases they classified. It conjointly
throws light into the importance of locally frequent
patterns and therefore the mining techniques used for
the purpose.
Int J Nano Corr Sci and Engg 3(3) (2016) 85-90
Editors: Dr S Rajendran, Dr M Pandiarajan, and A. Daniel Das
87
M. Chandamona et al.,
TABLE 1 – Summary of medical data mining techniques
REFERENCES
TECHNIQUES
UTILITY
[1],[2],[3],[4],[5],[6],[7],an
d [8]
Appriori and FP Growth
Associating rule mining for finding frequent
item sets(diseases) in medical databases
[9],[10]
Neural Networks
Extracting patterns, detecting trends
[9]
Genetic Algorithm
Classification of medical data
[11],[12]
Association Rule Mining
Finding frequent Patterns
[13]
Bayesian Ying Yang
Classification
DISEASE
Diabetic Diseases
Liver Diseases
(BYY)
[10],[14],[15],[16],[17],[18]
,
Decision Tree Algorithms such as
ID3,C4.5,C5,and CART
Decision Support
[21]
Outlier Prediction Technique
For improving classification accuracy
[22]
Fuzzy Cluster analysis
Analyzing medical images
[17],[23],[24]
Classification algorithm
Disease Classification
Cardio Vascular Diseases
[14],[25]
Bayesian Network algorihm
Modeling and analysis of medical data.
Coronary Heart Disease
[26]
Naive Bayesian
Improving Classification accuracy
Coronary Heart Disease
[28],[29]
Combined use of K-means SOM and
Naive Bayes
Medical diagnosis
[16]
Combination of SVM,ANN and ID3
Accurate Classification of medical data.
[30]
Clustering and Classification
Clustering and Classification Of biomedical
databases
[16],[31]
SVM
Disease Classification
[29]
Fuzzy Cognitive Maps
Drugs and Health effects classification
[24]
k-NN
Classification of Diseases
[19]
Diabetes
Diabetes , Cancer
Int J Nano Corr Sci and Engg 3(3) (2016) 85-90
Editors: Dr S Rajendran, Dr M Pandiarajan, and A. Daniel Das
88
M. Chandamona et al.,
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
Sarojini Balakrishnan (n.d). SVM Ranking with Backward
Search for Feature Selection in Type II Diabetes Databases.
IEEE. p1-6.
Arun K Pujari “Data Mining Techniques”, Edition 2001.
M. Ilayaraja Department of Computer Science &
Engineering Alagappa University Karaikudi, India
[email protected]. (2013). Mining Medical Data to
Identify Frequent Diseases using Apriori Algorithm. IEEE. 0
(0), p1-6.
J. C. Prather, D. F. Lobach, L. K. Goodwin, J. W.Hales , M.
L. Hage¸ W. Edward Hammond, “MedicalData Mining:
Knowledge Discovery in a Clinical DataWarehouse”, 1997.
HAI-BING MA, JIN ZHANG, YlNG-JIE FAN, YUN-FA W.
(2004). MINING FREQUENT PATTERNS BASED ON IS+TREE. IEEE. 0 (0), P1208-1213.
Goodwin L, Prather J, Schlitz K, Iannacchione My Hammond
W, Grzymala J, DataMining Issues for Impproved Birth
Outcomes, Biomed. Science Instrum, 34, 1997, pp. 291-296.
Evans S, Lemon S, Deters C, Fusaro R and Lynch H,
Automated Detection of hereditary Syndromes Using Data
Mining, Computers and Biomedical Research 30, 1997, pp.
337-348.
Krzysztof J. Cior , Medical Data Mining and Knowledge
Discovery. (n.d). From the guest Editor. IEEE. p1-2
Shusaku Tsumoto (n.d). Problems with Mining Medical
Data. IEEE. p1-2.
Syed Sibte Raza Abidi Kok Meng (n.d). Symbolic Exposition
of Medical Data-Sets: A Data Mining Workbench to
Inductively Derive Data-Defining Symbolic Rules. p1-6.
S. S. R. Abidi, K. M. Hoe, A. Goh, “Analyzing data clusters:
A rough set approach to extract cluster definingsymbolic
rules, Fisher, Hand, Hoffman, Adams (Eds.) Lecture Notes in
Computer Science: Advances inIntelligent Data Analysis, 4th
Intl. Symposium, IDA-01. Springer Verlag: Berlin, 2001.
Lionel Brunie, Maryvonne Miquel, Jean-Marc Pierson, and
Anne Tchounikine, “Information grids: managing and mining
semantic data in a grid infrastructure; open issues and
application to geno-medical data. 2003, 14th International
workshop on Database and Expert Systems Applications.
Wong Kok Seng, Rosli Bin Besar, Fazly Salleh Abas trosli,
(n.d). Collaborative Support for Medical Data Mining in
Telemedicine. IEEE. p1-6.
M. Ghannad-Rezaie, H. Soltanain-Zadeh, M.-R. Siadat, K.V.
Elisevich. (2006). Medical Data Mining using Particle
Swarm Optimization for Temporal Lobe Epilepsy. IEEE. p18.
Shusaku Tsumoto , (n.d). Problems with Mining Medical
Data. IEEE. p1-2.
Hidenao Abe AND Hideto Yokoi (n.d). Developing an
Integrated Time-Series Data Mining Environment for
Medical Data Mining. IEEE. p1-6.
Liu Jiquan Deng Wenliang Xudong Lu (n.d). Liu Jiquan
Deng Wenliang Xudong Lu Huilong Duan College of
Biomedical Engineering & Instrument Science Zhejiang
University Hangzhou 310027, China [email protected].
IEEE. p1-4.
Shusaku Tsumoto (n.d). Problems with Mining Medical Data.
IEEE. p1-2.
Hai Wang, Shouhong Wang 1. (n.d). Medical Knowledge
Acquisition through Data Mining. IEEE. 0 (0), p1-4.
[20] Gaurav N. Pradhan AND B. Prabhakaran (n.d).
ASSOCIATION
RULE
MINING
IN
MULTIPLE,
MULTIDIMENSIONAL TIME SERIES MEDICAL DATA.
IEEE. p1-4.
[21] Oliver Hogl, Michael Müller (2001). On Supporting Medical
Quality with Intelligent Data Mining. IEEE. p1-10.
[22] Umair Abdullah (2008). Analysis of Effectiveness of Apriori
Algorithm in Medical Billing Data Mining1. IEEE.p1-5.
[23] Cong-Rui Ji and Zhi-Hong Deng. (n.d). Mining Frequent
Ordered Patterns without Candidate Generation. IEEE. 0 (0),
P1-5.
[24] Hai-Tao He and Shi-Ling Zhang. (2007). A New method for
Incremental Updating Frequent patterns mining. IEEE. 0 (0),
p1-4.
[25] Carson Kai-Sang Leung∗ Christopher L. Carmichael and
Boyu Hao. (2007). Efficient Mining of Frequent Patterns
from Uncertain Data. IEEE. 0 (0), p489-494.
[26] Shariq Bashir, Zahid Halim, A. Rauf Baig. (2008). Mining
Fault Tolerant Frequent Patterns using Pattern Growth
Approach. IEEE. 0 (0), p172-179.
[27] Sunil Joshi and Dr. R. C. Jain. (2010). A Dynamic Approach
for Frequent Pattern Mining Using Transposition of
Database. IEEE. 0 (0), p498-501.
[28] Thanh-Trung Nguyen. (2010). An Improved Algorithm for
Frequent Patterns Mining Problem. IEEE. 0 (0), p503-507.
[29] Xiaoyong Lin and Qunxiong Zhu. (2010). Share-Inherit: A
novel approach for mining frequent patterns. IEEE. 0 (0),
p2712-2717.
[30] Markus Brameier and Wolfgang Banzhaf. (2001). A
Comparison of Linear Genetic Programming and Neural
Networks in Medical Data Mining. IEEE.p1-10.
[31] Doron Shalvi and Nicholas DeClaris., (n.d). An
Unsupervised Neural Network Approach to Medical Data
Mining Techniques. IEEE. 0 (0), p1-6.
[32] Adepele Olukunle and Sylvanus Ehikioya, (n.d). A Fast
Algorithm for Mining Association Rules in Medical Image
Data. IEEE. p1-7.
[33] Cindy L. Bethel and Lawrence O. Hall and Dmitry Goldgof
(n.d). Mining for Implications in Medical Data. IEEE. p1-4.
[34] Jeong-Yon Shim, Lei Xu (n.d). MEDICAL DATA MINING
MODEL FOR ORIENTAL MEDICINE VIA BYY BINARY
INDEPENDENT FACTOR ANALYSIS. IEEE. p1-4.
[35] Jenn-Lung Su, Guo-Zhen Wu, I-Pin Chao (2001). THE
APPROACH OF DATA MINING METHODS FOR
MEDICAL DATABASE. IEEE. p1-3.
[36] Safwan Mahmud Khan Md. Rafiqul Islam Morshed U. (n.d).
Medical Image Classification Using an Efficient Data Mining
Technique. IEEE, p1-6.
[37] Yanwei Xing, Jie Wang and Zhihong Zhao (2007).
Combination data mining methods with new medical data to
predicting outcome of Coronary Heart Disease. IEEE. p1-5.
[38] Tsang-Hsiang Cheng, Chih-Ping Wei, Vincent S. Tseng
(n.d). Feature Selection for Medical Data Mining:
Comparisons of Expert Judgment and Automatic Approaches
. IEEE. p1-6.
[39] Mohammad Saraee, George Koundourakis, Babis
Theodoulidis. (n.d). EASYMINER: DATA MINING IN
MEDICAL DATABASES. IEEE. p1-3.
[40] SAM CHAO, FAI WONG, “AN INCREMENTAL
DECISION
TREE
LEARNING
METHODOLOGYREGARDING
ATTRIBUTES
IN
MEDICAL DATA MINING”. Proceedings of the Eighth
Int J Nano Corr Sci and Engg 3(3) (2016) 85-90
Editors: Dr S Rajendran, Dr M Pandiarajan, and A. Daniel Das
89
M. Chandamona et al.,
[41]
[42]
[43]
[44]
[45]
International Conference on Machine Learning and
Cybernetics, Baoding, 12-15 July 2009.
My Chau Tu AND Dongil Shin (2009). A Comparative
Study of Medical Data Classification Methods Based on
Decision Tree and Bagging Algorithms. IEEE. p1-5.
Vili Podgorelec, Marjan HerikoMaribor, (n.d). Improving
Mining of Medical Data by Outliers Prediction. IEEE. p1-6.
Shuyan Wang Mingquan Zhou Guohua Geng (n.d).
Application of Fuzzy Cluster Analysis for Medical Image
Data Mining. IEEE. p1-6.
Asha Gowda Karegowda M.A.Jayaram (2009). Cascading
GA & CFS for Feature Subset selection in Medical Data
Mining. IEEE. p1-4.
Graduate Institute of Applied Information Sciences (2009).
MEDICAL DATA MINING USING BGA AND RGA FOR
WEIGHTING OF FEATURES IN FUZZY K-NN
CLASSIFICATION. IEEE. p1-6.
[46] Weimin Xue, Yanan Sun, Yuchang Lu (n.d). Research and
Application of Data Mining in Traditional Chinese Medical
Clinic Diagnosis. IEEE.p1-4.
[47] Ranjit Abraham, Jay B.Simha, Iyengar (n.d). A comparative
analysis of discretization methods for Medical Datamining
with Naïve Bayesian classifier. IEEE. p1-2.
[48] Syed Zahid Hassan and Brijesh Verma,(n.d). A Hybrid Data
Mining Approach for Knowledge Extraction and
Classification in Medical Databases. IEEE. p1-6.
[49] Michele Berlingerio (n.d). Mining Clinical Data with a
Temporal Dimension: a Case Study. IEEE. p1-8.
[50] Wojciech Froelich, Alicja Wakulicz-Deja (2009). Mining
Temporal Medical Data Using Adaptive Fuzzy Cognitive
Maps. IEEE. P1-8.
[51] Michael Barnathan, Jingjing Zhang, Vasileios (n.d). A WEBACCESSIBLE FRAMEWORK FOR THE AUTOMATED
STORAGE
AND
TEXTURE
ANALYSIS
OF
BIOMEDICAL IMAGES. IEEE. p1-3.
Int J Nano Corr Sci and Engg 3(3) (2016) 85-90
Editors: Dr S Rajendran, Dr M Pandiarajan, and A. Daniel Das
90